DRAMATIC PROJECT SUMMARY / ABSTRACT Multidrug-resistant tuberculosis (MDR-TB) was estimated to occur in 600,000 people in 2017. The roll-out of the GeneXpert? test has generated a substantial increase in the demand for MDR-TB treatment. However, current MDR-TB treatment regimens take 9 months or longer to complete and have substantial toxicity. Therefore, a shorter, less toxic treatment regimen is needed. We have designed a regimen that does not contain PZA or an injectable agent and limits the administration of linezolid to the initial 8 weeks of treatment, before the neuropathic side effects of linezolid occur. Animal studies support the likely efficacy of this regimen. Studies of fixed duration regimens to achieve treatment shortening are associated with high risk, since there are no validated ways to predict what duration of treatment will be optimal. We have developed an innovative Phase 2 study design (?duration-randomization?) to identify the shortest effective treatment duration. In this design, participants are randomized to four durations of treatment from the shortest to the longest likely effective duration. The results are then analyzed together to determine the optimal treatment duration. In the proposed multicenter, randomized, partially blinded, four-arm, phase 2 DRAMATIC Trial (Duration Randomized Anti-MDR-TB And Tailored Intervention Clinical Trial) we will examine the efficacy and safety of an all-oral regimen of bedaquiline, delamanid, levofloxacin, linezolid, and clofazimine given for 16, 24, 32 or 40 weeks. By modeling the results of the four durations together, the design achieves substantial statistical efficiency. The optimal treatment duration identified in this trial can then be validated in a larger prospective Phase 3 clinical trial. In addition, recent studies have demonstrated that baseline patient characteristics can predict TB treatment outcomes; we will therefore stratify participants into those with ?extensive? and those with ?non-extensive? disease to provide guidance for clinical treatment. Aim 1: To identify the optimal duration of an experimental MDR-TB treatment regimen consistent with a successful treatment outcome. Aim 2: To describe the relationship between baseline prognostic risk strata and successful treatment outcome. Aim 3: To establish that the rRNA synthesis ratio, a novel biologic marker based on M. tb precursor rRNA, is associated with relapse at the individual-level across the range of durations studied in the trial. Development of a shorter, better-tolerated treatment regimen will greatly enhance the ability of TB control programs to treat the growing number of patients. The DRAMATIC Trial will employ an innovative and efficient new design to establish a robust, nontoxic MDR-TB treatment regimen and identify the minimal duration for which it needs to be administered. These results will speed the process of moving forward to a confirmatory phase 3 clinical trial and increase the likelihood that such a trial is successful.